The invention relates to preparations of immunoreactive proteins and in particular in purified form and having a high percentage of immunoreactive molecules relative to the total number of molecules. These proteins can be obtained through fermentation in a fluidized reactor of a host cell, which is capable of expressing the immunoreactive protein and production of the protein from the host cell respectively from the culture medium used for culturing the cells. The protein preparations according to the invention are outstandingly suitable for producing diagnostic and therapeutic compositions.
Diagnostic and therapeutic proteins can be expressed in prokaryotic cells, such as for example E. coli (Houston et al., U.S. Pat. No. 5,132,405) as well as in eukaryotic cell systems, (for example Pichia pastoris, baby hamster kidney (BHK) cells, Chinese hamster Ovary (CHO-)cells, hybridomas, transgenic animals and plants (Meade et. al., U.S. Pat. No. 4,873,316) and can be purified by protein-chemical methods.
With prokaryotic cell systems, a cost efficient production of small carbohydrate-free proteins can be realized due to the simple culture media and the fast growth of the microorganisms in relatively simple fermentation systems.
However, for the production of complex, carbohydrate-carrying (glyco-) proteins, the afore-described eukaryotic cells systems have to be used which require complex fermentation systems and require the use of expensive culture media. The compositions of the glyco-protein preparation that are to be purified are being decisively influenced by the type of culture medium and the fermentation system used. In particular, differences of the glycosylation of the protein, which influences the half-life in the plasma as well as certain biological effector functions of the purified (glyco-) protein preparation are described in the respective technical literature. (Stahl et al., PNAS, 73 (1976), 4045-4049).
In addition to the afore-described influences on the carbohydrate composition, an essential factor which determines the quality of the product is the percentage share of functionally active (glyco-) proteins in the purified preparation, which can be also influenced by the type of cell media used, the conditions of fermentation and the purification methods applied.
As a preferred method used for determining the portion of functionally active radio-labeled molecules (for example Mab with physiological intact V-region) is the quantitative binding assay first described by Lindmo et al. (Immunological Methods 72 (1984) 77) in which an excess of antigen was used. This methods permits a quantitative determination of the immunoreactive and radioactive labeled Mab (monoclonal antibody) portion having a bonding capacity, after protein chemical purification. Other binding assays, such as the immunoreactivity test of Seitz et al. (European Journal of Nuclear Medicine 26 (1999), 1265) where immunoreactivities of up to 100% are described, permit only a relative determination of the immunoreactive portion of radio-labeled Mab in comparison to the unlabelled standard and are thus not suitable for determination of the absolute immunoreactive fraction. Similarly, the method described by Jagoda et al. (Journal of Immunological Methods 173 (1994), 191-201) directed to an affinity-chromatographic method for determining immunoreactivity has only limited applicability for a quantitative determination of the immunoreactive portion due to the relatively high portion of unspecific linkage with the affinity column (10-20%).
Likewise, the Lindmo-test utilized by most authors, and despite the authors claim that the method has general applicability, it is only suitable for the evaluation of radio-labeled Mab where binding can be realized to a distinct plateau by utilizing increasing amounts of antigen. Oftentimes, a distinct plateau is not reached with Mabs of low avidity or with a lack of cells that express high antigen densities per cell (Mattes, M. J., Int. J. Cancer: 61 (1995), 286-288) Even in investigations, where a definitive plateau can be reached, plotting the data from the saturation curves according to the Lindmo et al. method, produces two straight crossed lines, where the intersection with the Y-axis can lead to differing immunoreactivities. (Mattes, M. J., Int. J. Cancer: 61 (1995), 286-288, page 287, FIGS. 1 B and E). Thus, when using the Lindmo method the maximal specifically linked radioactive portion and the value extrapolated to the infinite antigen excess must be given. When the difference between these two values is >10%, then the extrapolated immunoreactivity value cannot be regarded as reliable (Mattes M. J., Int. J. Cancer: 61 (19915), 286-288, page 288).
With regard to the afore-described reservations concerning that method, when using the Lindmo test, the percent portion of the functionally active Mab molecule prepared with conventional fermenting methods, such as for example stirred fermentors and hollow fiber modules and purified with classical protein chemical methods which also comprise, besides protein A-columns, ion exchanger columns and in many cases also gel-chromatographic columns, is in the range of 40-80% (Mattes, M. J., Int. J. Cancer: 61 (1995), 286-288; Jagoda et al., Journal of Immunological Methods 173 (1994), 191-201; Morales-Morales et al., Nuclear Medicine & Biology, Vol. 26 (1999), 275-279; Boven et al., Blood, 67, N9 2 (1986), 429-435). Here, the radio-labeling method and the radioisotope used likewise have an important role.
This means that 20-60% of the Mab molecules of these prepared preparations do not carry out the desired diagnostic or therapeutic function due to their production method.
One exception known in the literature is the granulascint described by Schubinger et al., (Eur. J. Nucl. Med. 15: (1989), 605-608) where an immunoreactivity of 93 respectively 40% is described, depending on the respective labeling methods. However, this Mab was produced in the form of ascites (personal communication). This in vivo method fulfills neither economic nor modern regulatory requirements and therefor has to be regarded as an exception. Furthermore, it is noted, that the portion of the really specifically binding Mab molecules is smaller than the value determined in the Lindmo test. This is the result of the unspecific binding of radio-labeled Mab to e.g. the granulocytes used in an antigen excess even with the addition of a 10 000 times molar excess of unlabelled Mab (Schubinger et al., Euro J. Nucl. Med., 15 (1989), 605-608). Accordingly, after subtraction of unspecific binding the immunoreactivity value is at <90%.
In the case of radio-labeled diagnostic monoclonal antibodies, a reduced immunoreactivity of the Mab is a definite disadvantage, since radio-labeled molecules that are not binding to the specific target structure circulate in the blood, contribute to the unspecific background are partially absorbed by the liver and Mab the interpretation of nuclear-medical diagnosis more difficult. In the case of radio-labeled therapeutic antibodies, the portion of the Mab molecules not bound to the target tissue contributes to the unspecific and undesired irradiation of non-target tissue.
It is thus of great importance in particular for producing radio-labeled monoclonal antibodies to provide economically efficient in vitro methods that pass regulatory acceptance and with which production of Mab preparations that have an as high as possible immunoreactive portion can be realized.
At this time, highly immunoreactive portions can only be realized by means of complex immunoaffinity-chromatographic methods which are not suitable for subsequent commercialization by the pharma industry. Here, the functionally active molecules are separated from the functionally inactive molecules which are protein-chemically not distinguishable from the active molecules, through binding to antigen columns whereby they are also concentrated. Despite the relatively high cost in the production of the reagents necessary for the immunoaffinity chromatography and the high losses during purification, the portion of the functionally active Mab in such preparations utilized for research purposes is even <81%.